We wish to test the idea that subtypes of schizophrenia represent outcomes of different causes rather than variable responses to the same provocation. We intend to explore the distribution in a representative sample, and in the context of a family history of schizophrenia, of two variables: season of birth, and age at onset. I. Season of Birth: A. When compared with patients with schizophrenia who were born during the months of March-November, those born in December, January and February are: 1. less likely to have a poor premorbid history and have affected relatives 2. more likely to be younger at first admission and be born to a mother with a characteristic pattern of spring conception 3. to have had some obstetrical complication 4. differ in signs and symptoms II. Age at Onset: A. The genetic influence will vary according to age. l. There will be an inverse relationship between age at onset of schizophrenia (or age at first hospitalization) and the incidence of affected relatives, the severity of the disease B. The incidence of affected relatives will be particularly high when the index case is a female with age at onset below the median age for adults C. There will be a significant correlation for age at onset within families.